Google has had a significant spam problem for several monthsProblem. Search is also increasingly lacking in diversity. One of the biggest algorithm updates in company history was supposed to help. But reality speaks a different language. Has Google even lost control of its search?

According to a recent study from Germany, Google searches are getting worse and worse. The main reason for this: the increasing amount of search engine optimized but low-quality content. Since the rise of AI, this spam problem has become more and more apparent.

Fraudsters use artificial intelligence to create large amounts of content in order to appear high in Google searches and generate as many clicks as possible. They also flood their websites with advertisements. Facts play at best a subordinate role.

Google has a massive problem with AI spam

Google regularly makes algorithm adjustments to devalue bad content. However, such measures recently only had a temporary positive effect and appear to be more of a stopgap solution. Because spam websites are always finding new opportunities and ways to climb back up in the rankings and outsmart Google.

To finally put an end to the problem, Google recently carried out the most extensive algorithm update in its company history. The so-called “March 2024 Core Update” ran for a full 45 days. On May 5, 2024 the company declared it over.

But Google still doesn't seem to have mastered the situation. Because numerous SEO experts have found that AI spam continues to occupy top positions in search results. Lily Ray, vice president of search strategy and research at digital marketing agency Amsive, tells Tech magazine Fast Company in addition:

I've never seen so many big companies get away with publishing really low-quality content that appears in the top spots. Every day I probably get 20 to 30 desperate messages because their websites are either being overtaken by spam or AI content, or because their content is being exploited and plagiarized by AI tools.

Has Google lost control of its search?

Google generates over 75 percent of its revenue from selling ads on its search engine results pages – also called SERPs. However, this revenue depends on how trustworthy and useful the company's search results are.

But if Google searches get worse and worse, not only users but also advertisers could leave. Taking action against AI spam should actually be in the company's interest. But Google is, in a sense, shooting itself in the foot.

The reason: It has never been so easy to outsmart the Google algorithm. If you enter the query “SEO with AI” into a search engine, numerous instructions appear on how you can litter the Internet with low-quality content in order to profit in the form of clicks and advertising revenue.

Ironically, there are even videos on how to use Gemini, Google's in-house AI, to achieve top search rankings. In order to combat AI spam, Google has been favoring content that is likely to come from humans for some time now.

AI spam: Google is tripping itself up

According to a current analysis, the social news aggregator Reddit and the question-answer network Quora have recently dominated the so-called SERP results for “discussions and forums” by a significant margin. The problem is that Google searches are increasingly lacking in diversity.

According to the analysis, many Reddit threads that appear in the SERP results also contain completely meaningless and thematically inappropriate content, which was obviously formulated by an AI. The creators of such posts only care about including affiliate links in the posts.

Google is also currently experimenting with a new feature called Search Generative Experience (SGE). This is short information based on the Google AI Gemini and occasionally appears at the top of search results. The function is still in the testing phase, but it is already clear that Google is desperately trying to integrate AI into its search, but seems to be powerless against AI spam.

It is commendable that the SGE results, unlike ChatGPT, contain source information, but this also ensures that Google reduces data traffic to other websites – at the expense of advertising revenue. The idea behind it still makes sense – but only if Google gets its spam problem under control.

Can Google take back control of its search?

By integrating the in-house Gemini AI, Google wants to secure its position as the leading search engine, as more and more users are now using platforms like TikTok or tools like ChatGPT to search for information on the Internet.

This is laudable in that this development stimulates competition. However, Google is faced with a dilemma. Because the company must either rely on artificial intelligence or risk losing market power. In order to be able to play in the AI ​​game, the search engine giant must, above all, ensure that AI spam is contained.

The question, however, is whether Google is even capable of doing this. Because the past algorithm updates were at best an emergency solution. However, the fact that there are numerous instructions circulating in Google searches that explain how to outsmart the in-house algorithms raises doubts. In other words: Either Google is unable to take action against AI spam or has become too dependent on advertising revenue.

Also interesting:


Leave a Reply

Your email address will not be published. Required fields are marked *